The Digital Bits: grain is not a defect on the disc! - Page 7

Thats why you can do whatever you want to with your own display, but dont make the studio DNR my movies.

But unless you are going to watch it in analog film in your home theater, it will have to be converted to digital anyway. And when you convert, you will lose quality in the transfer, won't you? It is like trying to make your photographs into digital photos. They can look great, but something always ends up "off".

So if we go with your argument (which I disagree with), you have a choice. Would you rather see a better picture in the theater or your home theater knowing that you would have to choose one or the other?

I think you need to get under that "tree of no screen grain" and out of the sun.

Not really. Because you can in post control the conversion. Its like when the studios do their 10bit to 8bit conversion. If the do it wrong you suddenly gets banding, and if you do it right its looks almost as good as the source.

But if the cam shoot the movie in 8bits from the beginning you dont have the same rome to play with the material. So even if BD cant go over 8bits, you still have advantage of shooting/scanning with 10 or 12 bits.

Look on a movie like Apocalypto, I can see on the BD that it was shoot on Video and not film. So there is something more to the footage then just resolution thats still gets captured by the BD medium.

But Im not against that digital cams will be used in filmproductions. Just that its not better then film on all apsects of PQ.

I think it should give up to 4 times picture resolution, especially on films that have been rendered using CGI at that resolution (especially if it is transferred direct from digital to a Blu-ray disc and viewed on a 4k display device).

Though the visual difference will be less on films that use soft focus lenses or aren't totally in focus, etc.

LOL Thats pretty good.

I do think that the transition is inevitable. I may take some time, but the economics favors digital capturing and then viewing.

So here is another question. How do you feel about artifical grain?

Transformer on HD DVD is AVC.

Good call. All the rabid Transformers kids on Blu-ray.com got me thinking it was VC-1 with their cries for a new AVC encode. So I guess it's the reverse, Bay preferred the VC-1 encode.

Transformers looks great on HD DVD. I personally like VC-1 better. Not so much quality reasons, but it is more compact and easier to work with on my HTPC. I think all the formats are just fine though. It usually comes down to how they did the transfer anyway and not so much the codec.... unless it was shot digitally of course

I got my copy the same way the autobots found the allspark..... EBAY! lol

Can't wait to get Iron Man

Crank was shot completely digital @ 1080p/24. maybe they cleaned it up post? I don't know or care, as it is one of the sharpest and most detailed BD movies I own.

Detail is neither loose nor tight.

http://loseloose.com/

Good one. I think before anyone is allowed to post on the internet, they need to take a test showing they know the difference between lose and loose.

That and a myriad of others . . . I thought 'to' and 'too' and 'your' and 'you're' were covered pretty early

I think it's very easy to get bad results with the RED if you don't understand the RAW concept and how to expose for RAW. It's not how you expose film and it's not how you expose traditional HD cameras. In addition RED has not provided an easy interface so far to see clearly what's going on on the RAW level before curves and metadata is applied to the monitoring output. Apparently that misleads many into chosing wrong shooting parameters and/or grade versions already screwed up by wrong selection of curves, ISO etc. and then pulling their hair because the apparent latitude is 6 stops instead of 8 or 9 or even more. RED is working frantically on firmware 16 which is supposed to make it much easier to see what you get from the sensor and select your exposure, lighting and filters for best results. I know RED can look fantastic. I have seen it. It can also look like crap if you insist (on purpose or out of lack of experience) on using it that way.
Soderbergh's films are in Cannes. They should give a good indication what you could do a year ago.

They are having problems associated with the rolling shutter : not necessary latitude. The people using it are extremely experienced.

Part of Crank was also shoot on a SD cam.

Digital cams works abit differently then film (in case anyone didnt know )

They have their image sensors they read from then the incamera processing takes on for the readout. So the camera manufacturers tweaks the cam before it goes to tape/hardrive.

So its possible to tweak the cam and get very low noise, but in return you lose on the other pq qualites.

So the manufacturers must make compromisses on what makes the best final image.

But the main camera for Crank is the F950 and its a very good cam. So its not strange it can produce good results. But I havnt seen any screenshoot from Crank that rivals film in any way.

It's telling that you are talking about entertainment and entertainment only. Do you think a football game and any feature film is essentially the same and exchangeable?

Then they should use a CCD based cam instead of a CMOS.


Viper and F950 should be better options.

They are not shooting anything themselves. They are dealing with footage supplied by a client. They will likely tell them to reshoot it, what they shoot it on is the client's business. I suspect they will end up using 35mm ( as it works).

Oh good one! At least I know the difference between film grain and digital Noise artifacts..which btw is evident in this thread.

Wrong!
The Gamma transfer function and sensitivity is "user defined" variable for even standard ENG cameras and most certainly IS a funamental tool on all state of the art camera designs.
http://media.panavision.com/Referenc.../f23_broch.pdf
S-LOG Gamma (LUT)

Its apparent you don’t understand fundamental digital processing.

I love when folks look resort to rip others apart by insulting them or degrading them. Shows alot of class. I hope ripping me makes you feel better.

I dont think they are interchangeable..completely. But they are both meant to be entertainment.

I am very much into art as well. Have many prints as I can't afford the real ones. Visit museums. Have taken a multitude of art classes and paint in my spare time.

I consider some films art. Not all. but some. 2001. Dune. The Matrix. A different type of art... a combination between a narrative and a visual with both being equally important.

That aside, I also see them as entertainment. Books can be entertaining as well as artistic. So can traditional art. I prefer Van Gogh. Sure, there is more realism in many other artist, but I have always been fascinated in the raw passion of Van Gogh. That being said. Van Gogh and DaVinci used the same medium for the most part. Sure you can use different types of paint and substrates, but they are both very similar in basic structure.

So while a football game and film are different, they are somewhat comparable. They do show games now in theaters too you know. And many films are made that are intended for TVs as much as they are big screens.

When I read a good book, I get lost in reality and forget that I am just reading a book and imagine that I am in the story. I do the same in a movie. don't you? To many, screen grain is now being seen as a distraction rather than adding to the effect. So, I as well as others prefer the digital without screen grain.

I don't understand why you feel it necessary to bash us. And blindly state that film is better. And digital is so worse. Obviously, screen grain does not bother you, correct? So film may look better to you.

Here is a great example. most will agree that DLP looks better than LCD projection especially when it comes to contrast. But if you see the DLP rainbow, it is virtually unwatchable. So even though DLP may technically be a better option in PQ, it is not for those affected. There are only a small percentage affected by DLP rainbows, but there seems to be a higher number who dislike screen grain. atleast there seems to be more to me from all the posts that I have read.

Film technology is a powerful esthetic tool for directors. It has unmatched exposure latitude (aka wide dynamic range) and a seemingly infinite color palette to work with.

Just because those shinny disc we call DVD's apparently have great difficultly compressing the obligatory fundamental characteristics of film (aka grain) is in the end just a limitation of the codec and the digital medium itself.

Debating art and the sublime is pointless!

I am not talking about limitations of DVD here. You can see grain the screen itself. It is not something made up by compression. It is something you can see. Are the spots where you can see grain accurately relating the color infinitely?

You are right about the color palette. But even that can be debated. and the colors are changed and distorted anyway in film during editing. I don't know the number of colors that bluray alone can render but I imagine it is in the millions. From what I have read, even the new laser DLP projectors coming out can only render up to 80% of our viewable spectrum anyway.

Well since the CCD readout is analog you do have some preprocessing during the AD conversion. Sure some cams give you more control then the other, but not every tweak is adjustable from the camera menu.

There is more to the image then adjusting the gammacurves.

While this is not pulled from an article, this poster had an excellent point about the limitations of film.


Another thing to take into account in the current reality of fillmaking is that the film itself is scanned into digital form in order to add CGI and edit the film. They no longer cut and splice film in the editing room to make most movies. After the editing, it is then transferred back to film. This will inherently create artifacts even in the smalles scale. So in reality, film is not infinite in the purist sense.

You may argue that scanning is better than raw digital capture. You may be right right, and you may be wrong as there is alot that goes into the transfer and all things ar enot equal. There can even be differences between individual film scanners in the same model let alone different models.

What it all comes down to is the perceived difference. Most people can see a difference between 480 and 720p once you get higher than 27". Most can see a difference between 720p and 1080p. How much of a difference can one see between 1080 and 2000? You can argue that the "big screen" is bigger, but many of us have screens where the viewing angle (which is what is important) is actually greater than what most watch in the theater. Most people seem to sit in the back of the theater.

In my home theater, I have a viewing angle of 40 degrees. When I go to the theater, I sit about midway and estimate a viewing angle of about 32. I imagine many toward the back have a viewing angle in the 20s.

at 40 degrees, I can notice a difference obviously between 480p and 720. I can also between 720 and 1080. I may notice a difference between 2K and 4K, but that is questionable. Would any of us really see a difference between 4K and 6K, 4K and 8K. And that is 40 degrees. At 28 degrees, I doubt there would really be any perceived difference.

And yes, I know there is more than resolution.... but resolution is none the less extremely important, isnt it?

So why introduce screen grain bits or allow it to exist to provide, in theory, a resolution that is not perceivable by most? That is the essence of my argument. To me, HD means that the image is cleaner and crisper. That is what I am looking for. It doesn't make it look plastic like... just more crisp. Just like a new car looks all beautiful and shiny while a 4 year old car can look a little more dull. Sure most cars look a little worn, but it doesn't make a new shiny car look less real.

As I stated above, image "detail" is embedded in the grain pattern for each frame and this pattern varies frame to frame.

The presence of noise (random grain) in an image severely degrades compression efficiency. In MPEG, for example, noise degrades interframe compression by adversely impacting the performance of "motion estimation," which achieves interframe compression by cross-referencing matching blocks of pixels in neighboring frames. In motion estimation, noise interferes with the identification of such matching blocks. In addition, noise also affects intraframe compression by reducing the correlation among neighboring pixel values, thereby reducing the compression efficiency achieved by quantizing video data transformed under a discrete cosine transform.

10bit log 4k in and out I would say is indistinguishable from the original camera neg in all the areas that matter. Has been since about 1994.

Seeing that these are digitally scanned and this will happen regardless, then why not avoid the film step which introduces this degredation in the first place?

Okay that's it for me . I 'm not wasting any more time trying to explain these issues to someone who can't even be bothered to make a stab at understanding the fundamentals.

And that is the real debate then isn't it. Is the 4K scanned image distinguishable from the 4K raw image using a digital camera? And if so at what angle?

I understand that you can in theory get a better transfer scanning film... at the moment, but would this difference be perceivable at 25 degrees, 30 degrees, 35 degrees, 40 degrees?

Is this increased better transfer worth introducing noticeable image degradation with the introduction of scanned screen grain if the difference in resolution and clarity is not a perceivable difference from a 25-35 degree angle?

It is a trade off in the end. The question is which poison do you prefer?

I think we could all agree that either method would be indistinguishable when scaled down to 1080p, right? If this is true then the digitally captured image would looked cleaner on our home theater screens, wouldn't it?